kernel_optimize_test/arch/x86/mm/srat_32.c

287 lines
8.1 KiB
C
Raw Normal View History

/*
* Some of the code in this file has been gleaned from the 64 bit
* discontigmem support code base.
*
* Copyright (C) 2002, IBM Corp.
*
* All rights reserved.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
* NON INFRINGEMENT. See the GNU General Public License for more
* details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*
* Send feedback to Pat Gaughen <gone@us.ibm.com>
*/
#include <linux/mm.h>
#include <linux/bootmem.h>
#include <linux/memblock.h>
#include <linux/mmzone.h>
#include <linux/acpi.h>
#include <linux/nodemask.h>
#include <asm/srat.h>
#include <asm/topology.h>
#include <asm/smp.h>
#include <asm/e820.h>
/*
* proximity macros and definitions
*/
#define NODE_ARRAY_INDEX(x) ((x) / 8) /* 8 bits/char */
#define NODE_ARRAY_OFFSET(x) ((x) % 8) /* 8 bits/char */
#define BMAP_SET(bmap, bit) ((bmap)[NODE_ARRAY_INDEX(bit)] |= 1 << NODE_ARRAY_OFFSET(bit))
#define BMAP_TEST(bmap, bit) ((bmap)[NODE_ARRAY_INDEX(bit)] & (1 << NODE_ARRAY_OFFSET(bit)))
/* bitmap length; _PXM is at most 255 */
#define PXM_BITMAP_LEN (MAX_PXM_DOMAINS / 8)
static u8 __initdata pxm_bitmap[PXM_BITMAP_LEN]; /* bitmap of proximity domains */
#define MAX_CHUNKS_PER_NODE 3
#define MAXCHUNKS (MAX_CHUNKS_PER_NODE * MAX_NUMNODES)
struct node_memory_chunk_s {
unsigned long start_pfn;
unsigned long end_pfn;
u8 pxm; // proximity domain of node
u8 nid; // which cnode contains this chunk?
u8 bank; // which mem bank on this node
};
static struct node_memory_chunk_s __initdata node_memory_chunk[MAXCHUNKS];
static int __initdata num_memory_chunks; /* total number of memory chunks */
static u8 __initdata apicid_to_pxm[MAX_APICID];
int acpi_numa __initdata;
static __init void bad_srat(void)
{
printk(KERN_ERR "SRAT: SRAT not used.\n");
acpi_numa = -1;
num_memory_chunks = 0;
}
static __init inline int srat_disabled(void)
{
return numa_off || acpi_numa < 0;
}
/* Identify CPU proximity domains */
void __init
acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *cpu_affinity)
{
if (srat_disabled())
return;
if (cpu_affinity->header.length !=
sizeof(struct acpi_srat_cpu_affinity)) {
bad_srat();
return;
}
if ((cpu_affinity->flags & ACPI_SRAT_CPU_ENABLED) == 0)
return; /* empty entry */
/* mark this node as "seen" in node bitmap */
BMAP_SET(pxm_bitmap, cpu_affinity->proximity_domain_lo);
x86, acpi: Parse all SRAT cpu entries even above the cpu number limitation Recent Intel new system have different order in MADT, aka will list all thread0 at first, then all thread1. But SRAT table still old order, it will list cpus in one socket all together. If the user have compiled limited NR_CPUS or boot with nr_cpus=, could have missed to put some cpus apic id to node mapping into apicid_to_node[]. for example for 4 sockets system with 64 cpus with nr_cpus=32 will get crash... [ 9.106288] Total of 32 processors activated (136190.88 BogoMIPS). [ 9.235021] divide error: 0000 [#1] SMP [ 9.235315] last sysfs file: [ 9.235481] CPU 1 [ 9.235592] Modules linked in: [ 9.245398] [ 9.245478] Pid: 2, comm: kthreadd Not tainted 2.6.37-rc1-tip-yh-01782-ge92ef79-dirty #274 /Sun Fire x4800 [ 9.265415] RIP: 0010:[<ffffffff81075a8f>] [<ffffffff81075a8f>] select_task_rq_fair+0x4f0/0x623 ... [ 9.645938] RIP [<ffffffff81075a8f>] select_task_rq_fair+0x4f0/0x623 [ 9.665356] RSP <ffff88103f8d1c40> [ 9.665568] ---[ end trace 2296156d35fdfc87 ]--- So let just parse all cpu entries in SRAT. Also add apicid checking with MAX_LOCAL_APIC, in case We could out of boundaries of apicid_to_node[]. it fixes following bug too. https://bugzilla.kernel.org/show_bug.cgi?id=22662 -v2: expand to 32bit according to hpa need to add MAX_LOCAL_APIC for 32bit Reported-and-Tested-by: Wu Fengguang <fengguang.wu@intel.com> Reported-by: Bjorn Helgaas <bjorn.helgaas@hp.com> Tested-by: Myron Stowe <myron.stowe@hp.com> Signed-off-by: Yinghai Lu <yinghai@kernel.org> LKML-Reference: <4D0AD486.9020704@kernel.org> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2010-12-17 11:09:58 +08:00
/* don't need to check apic_id here, because it is always 8 bits */
apicid_to_pxm[cpu_affinity->apic_id] = cpu_affinity->proximity_domain_lo;
printk(KERN_DEBUG "CPU %02x in proximity domain %02x\n",
cpu_affinity->apic_id, cpu_affinity->proximity_domain_lo);
}
/*
* Identify memory proximity domains and hot-remove capabilities.
* Fill node memory chunk list structure.
*/
void __init
acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *memory_affinity)
{
unsigned long long paddr, size;
unsigned long start_pfn, end_pfn;
u8 pxm;
struct node_memory_chunk_s *p, *q, *pend;
if (srat_disabled())
return;
if (memory_affinity->header.length !=
sizeof(struct acpi_srat_mem_affinity)) {
bad_srat();
return;
}
if ((memory_affinity->flags & ACPI_SRAT_MEM_ENABLED) == 0)
return; /* empty entry */
pxm = memory_affinity->proximity_domain & 0xff;
/* mark this node as "seen" in node bitmap */
BMAP_SET(pxm_bitmap, pxm);
/* calculate info for memory chunk structure */
paddr = memory_affinity->base_address;
size = memory_affinity->length;
start_pfn = paddr >> PAGE_SHIFT;
end_pfn = (paddr + size) >> PAGE_SHIFT;
if (num_memory_chunks >= MAXCHUNKS) {
printk(KERN_WARNING "Too many mem chunks in SRAT."
" Ignoring %lld MBytes at %llx\n",
size/(1024*1024), paddr);
return;
}
/* Insertion sort based on base address */
pend = &node_memory_chunk[num_memory_chunks];
for (p = &node_memory_chunk[0]; p < pend; p++) {
if (start_pfn < p->start_pfn)
break;
}
if (p < pend) {
for (q = pend; q >= p; q--)
*(q + 1) = *q;
}
p->start_pfn = start_pfn;
p->end_pfn = end_pfn;
p->pxm = pxm;
num_memory_chunks++;
printk(KERN_DEBUG "Memory range %08lx to %08lx"
" in proximity domain %02x %s\n",
start_pfn, end_pfn,
pxm,
((memory_affinity->flags & ACPI_SRAT_MEM_HOT_PLUGGABLE) ?
"enabled and removable" : "enabled" ) );
}
/* Callback for SLIT parsing */
void __init acpi_numa_slit_init(struct acpi_table_slit *slit)
{
}
void acpi_numa_arch_fixup(void)
{
}
/*
* The SRAT table always lists ascending addresses, so can always
* assume that the first "start" address that you see is the real
* start of the node, and that the current "end" address is after
* the previous one.
*/
static __init int node_read_chunk(int nid, struct node_memory_chunk_s *memory_chunk)
{
/*
* Only add present memory as told by the e820.
* There is no guarantee from the SRAT that the memory it
* enumerates is present at boot time because it represents
* *possible* memory hotplug areas the same as normal RAM.
*/
if (memory_chunk->start_pfn >= max_pfn) {
printk(KERN_INFO "Ignoring SRAT pfns: %08lx - %08lx\n",
memory_chunk->start_pfn, memory_chunk->end_pfn);
return -1;
}
if (memory_chunk->nid != nid)
return -1;
if (!node_has_online_mem(nid))
node_start_pfn[nid] = memory_chunk->start_pfn;
if (node_start_pfn[nid] > memory_chunk->start_pfn)
node_start_pfn[nid] = memory_chunk->start_pfn;
if (node_end_pfn[nid] < memory_chunk->end_pfn)
node_end_pfn[nid] = memory_chunk->end_pfn;
return 0;
}
int __init get_memcfg_from_srat(void)
{
int i, j, nid;
if (srat_disabled())
goto out_fail;
if (num_memory_chunks == 0) {
printk(KERN_DEBUG
"could not find any ACPI SRAT memory areas.\n");
goto out_fail;
}
/* Calculate total number of nodes in system from PXM bitmap and create
* a set of sequential node IDs starting at zero. (ACPI doesn't seem
* to specify the range of _PXM values.)
*/
/*
* MCD - we no longer HAVE to number nodes sequentially. PXM domain
* numbers could go as high as 256, and MAX_NUMNODES for i386 is typically
* 32, so we will continue numbering them in this manner until MAX_NUMNODES
* approaches MAX_PXM_DOMAINS for i386.
*/
nodes_clear(node_online_map);
for (i = 0; i < MAX_PXM_DOMAINS; i++) {
if (BMAP_TEST(pxm_bitmap, i)) {
int nid = acpi_map_pxm_to_node(i);
node_set_online(nid);
}
}
BUG_ON(num_online_nodes() == 0);
/* set cnode id in memory chunk structure */
for (i = 0; i < num_memory_chunks; i++)
node_memory_chunk[i].nid = pxm_to_node(node_memory_chunk[i].pxm);
printk(KERN_DEBUG "pxm bitmap: ");
for (i = 0; i < sizeof(pxm_bitmap); i++) {
printk(KERN_CONT "%02x ", pxm_bitmap[i]);
}
printk(KERN_CONT "\n");
printk(KERN_DEBUG "Number of logical nodes in system = %d\n",
num_online_nodes());
printk(KERN_DEBUG "Number of memory chunks in system = %d\n",
num_memory_chunks);
for (i = 0; i < MAX_APICID; i++)
apicid_2_node[i] = pxm_to_node(apicid_to_pxm[i]);
for (j = 0; j < num_memory_chunks; j++){
struct node_memory_chunk_s * chunk = &node_memory_chunk[j];
printk(KERN_DEBUG
"chunk %d nid %d start_pfn %08lx end_pfn %08lx\n",
j, chunk->nid, chunk->start_pfn, chunk->end_pfn);
if (node_read_chunk(chunk->nid, chunk))
continue;
memblock_x86_register_active_regions(chunk->nid, chunk->start_pfn,
min(chunk->end_pfn, max_pfn));
}
x86: Fix checking of SRAT when node 0 ram is not from 0 Found one system that boot from socket1 instead of socket0, SRAT get rejected... [ 0.000000] SRAT: Node 1 PXM 0 0-a0000 [ 0.000000] SRAT: Node 1 PXM 0 100000-80000000 [ 0.000000] SRAT: Node 1 PXM 0 100000000-2080000000 [ 0.000000] SRAT: Node 0 PXM 1 2080000000-4080000000 [ 0.000000] SRAT: Node 2 PXM 2 4080000000-6080000000 [ 0.000000] SRAT: Node 3 PXM 3 6080000000-8080000000 [ 0.000000] SRAT: Node 4 PXM 4 8080000000-a080000000 [ 0.000000] SRAT: Node 5 PXM 5 a080000000-c080000000 [ 0.000000] SRAT: Node 6 PXM 6 c080000000-e080000000 [ 0.000000] SRAT: Node 7 PXM 7 e080000000-10080000000 ... [ 0.000000] NUMA: Allocated memnodemap from 500000 - 701040 [ 0.000000] NUMA: Using 20 for the hash shift. [ 0.000000] Adding active range (0, 0x2080000, 0x4080000) 0 entries of 3200 used [ 0.000000] Adding active range (1, 0x0, 0x96) 1 entries of 3200 used [ 0.000000] Adding active range (1, 0x100, 0x7f750) 2 entries of 3200 used [ 0.000000] Adding active range (1, 0x100000, 0x2080000) 3 entries of 3200 used [ 0.000000] Adding active range (2, 0x4080000, 0x6080000) 4 entries of 3200 used [ 0.000000] Adding active range (3, 0x6080000, 0x8080000) 5 entries of 3200 used [ 0.000000] Adding active range (4, 0x8080000, 0xa080000) 6 entries of 3200 used [ 0.000000] Adding active range (5, 0xa080000, 0xc080000) 7 entries of 3200 used [ 0.000000] Adding active range (6, 0xc080000, 0xe080000) 8 entries of 3200 used [ 0.000000] Adding active range (7, 0xe080000, 0x10080000) 9 entries of 3200 used [ 0.000000] SRAT: PXMs only cover 917504MB of your 1048566MB e820 RAM. Not used. [ 0.000000] SRAT: SRAT not used. the early_node_map is not sorted because node0 with non zero start come first. so try to sort it right away after all regions are registered. also fixs refression by 8716273c (x86: Export srat physical topology) -v2: make it more solid to handle cross node case like node0 [0,4g), [8,12g) and node1 [4g, 8g), [12g, 16g) -v3: update comments. Reported-and-tested-by: Jens Axboe <jens.axboe@oracle.com> Signed-off-by: Yinghai Lu <yinghai@kernel.org> LKML-Reference: <4B2579D2.3010201@kernel.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-12-16 09:59:02 +08:00
/* for out of order entries in SRAT */
sort_node_map();
for_each_online_node(nid) {
unsigned long start = node_start_pfn[nid];
unsigned long end = min(node_end_pfn[nid], max_pfn);
memory_present(nid, start, end);
node_remap_size[nid] = node_memmap_size_bytes(nid, start, end);
}
return 1;
out_fail:
printk(KERN_DEBUG "failed to get NUMA memory information from SRAT"
" table\n");
return 0;
}